Investors are much more comfortable with parametric or objective triggers explains John Ryan.

In recent years, there has been a growth in the use of capital markets to finance catastrophic exposure for natural perils. It is supplementary to conventional reinsurance rather than replacing it and in most cases fulfils a need that for one reason or another cannot easily be met by the existing markets. Capital market instruments have generally been confined to catastrophic, easily definable events. Consequently, coverage or legal dispute is normally not an issue.

They are also used for remote events. Probabilities of loss are less than 1 in 20 years and usually significantly less. The contracts have usually been for short terms, but a number have been longer than conventional reinsurance contracts, and there is a trend for longer term contracts. That allows the issue cost to be analysed over a longer period of time and provides continuity of coverage.

A recent development has been the use of weather derivatives. They are quoted on the London International Financial Futures and Options Exchange (LIFFE) as well as a number of over-the-counter (OTC) markets. The basis of these contracts is that they pay out in the event that the weather is different from that assumed. The structure is essentially the same as a put or call option. Again, their success depends on objective triggers and the fact that sufficient statistics are available to price the derivative. These are often not for catastrophic events, but it is important to recognise that the trigger is objective. It depends on the weather event occurring or not occurring and is not related to indemnification.

A feature of recent developments has been the introduction of parametric triggers for catastrophe bonds, making them much more akin to weather derivatives. Again, instead of using an indemnification trigger, the bond pays out on the basis of an event occurring. The amount of the payment is defined by way of a parameter. This may simply be the size of an earthquake weighted within a particular geographic area, or it may require the modelling firm to estimate what potential damage would occur in various windstorms and calculate an appropriate index. The point is that the payout can be much more swift and is not tied into the settlement of claims or disputes about quantum. Therefore, it can help with liquidity issues.

Basis risk
Weather securitisation can be written on a physical event basis, but the trend is very much more towards using a parametric approach. Clearly, with a parametric trigger the purchaser of the contract may have a basis risk in terms of what is being covered and the requirements. This means the contract may not meet the purchaser's losses. In many cases, therefore, success in completing the transaction is in ensuring that the insured is comfortable with the amount of basis risk. To reach this point requires a great degree of confidence in the modelling that goes in to the underlying exposures and their correlations with the event, for example a brewery and loss of sales in a wet summer.

This is an area where actuaries and other modellers are increasingly becoming involved. Indeed, in many cases the costs of placing the business may be dwarfed by the cost of modelling the basis risk.

Why accept basis risk? The advantage of accepting basis risk on the insured's part is that money becomes immediately payable. There are no disputes over the amounts involved or, indeed, any issues of claims settling. Where losses must be paid upfront to trust funds, for instance for Lloyd's and London market companies reinsuring US insurers, then liquidity can be immediately available. Nor are there difficulties with the auditors over at the year-end as to whether the insurance will respond or not. Furthermore, as the insured and not the insurer accepts the basis risk, there is a risk margin that the insured can save on the estimates of the cost of the damage of the underlying covers.

This also allows access to new markets. Investors are much more comfortable with parametric or objective triggers than with the assumption that insurers will handle claims fairly. Conventional reinsurance deals with this issue because reinsurers understand the claims settlement process. Investors do not. On the other hand, even if the insured is comfortable with the basis risk, there can be a degree of flexibility once the contract is issued if the basis risk changes, although investors may require additional compensation for renegotiating.

Bonds are usually rated by the conventional rating agencies, as this is the approach that most investors are comfortable with. The basis of the rating is a function of the coverage and the likelihood of the event, and specialist modelling firms provide the analysis in the same way they do for conventional reinsurance. Nevertheless, however good a modelling firm is, there will still be `modelling error', and investors will wish to reflect that in the price. Conventional reinsurers are more comfortable with this type of risk, thus bonds may appear more expensive than reinsurance. In reality, the difference is often only the modelling risk.

On the other hand, since the risk is non-correlated with the rest of the investor's business, it is not necessary to charge a premium for utilisation of scarce capacity. This applies to weather derivatives where most of the players entering the market are part of larger, thus there is a large degree of non-correlation.

Belief in the modelling is, therefore, important not only for pricing the bond but also for assessing basis risk. For actual underwriting purposes, many reinsurers will use a number of different firms and make their own assessments. However, for investment purposes, the reputation of the modelling firm is paramount. Investors are much more familiar with to basing their decisions on the reputation of the firms concerned rather than a detailed knowledge of the science.

Role of the broker
An insurance broker often initiates the transaction. The broker is able to relate the cover to alternatives and advise on structures and can handle the relationships with the modelling firms and co-ordinate the whole project. Alternatively, some of the investment banks or reinsurers may wish to do this.

Investment bankers need to handle the relationships with the investors and deal with capital market structures. The broker, however, will develop appropriate relationships with the bankers and may also deal with any necessary warehousing business, i.e. placing several transactions with the bank in order to make it to commercial size to take into the investment markets.

The future
These types of transactions will not replace conventional reinsurance. Conventional reinsurance responds well and has served the insurance industry well over the years. However, these contracts will be increasingly common, particularly in parametric situations. They can be used to cover extremely remote events for which the capacity may not be available in the conventional market. They deal with the contingent credit problem on reinsurance and they also handle the contingent liquidity aspect in that the loss becomes immediately payable and there is no need to wait for the reinsurers to pay. Furthermore, if the funds are available up front there is no credit risk in the event of the loss being excessively high.

Our experience is that we are seeing a much greater involvement in these types of financing arrangements for governments, quasi government agencies and Third World projects, where there is a real need for finance in the case of the adverse event happening.

By John Ryan

John Ryan is Managing Director of hybrid solutions at Heath Lambert in London. He is also a member of the Institute of Actuaries. Email: